Re: addsuffix manual example wrong, se_implicit test 4 fails.

[John E. Malmberg]

On 9/18/2014 8:50 AM, Paul Smith wrote:
> On Wed, 2014-09-17 at 20:34 -0500, John E. Malmberg wrote:
> It works fine for me in GNU make 4.0 on Linux Mint 17 (basically Ubuntu
> 14.04) (actually this is from the current Git head but I'm pretty sure
> nothing in this area has changed since the 4.0 release):
>
>    $ mkdir /tmp/foo /tmp/bar
>    $ touch /tmp/foo/foo.c /tmp/bar/foo.c
>    $ touch foo.h
>
>    $ cat Makefile
>    .SECONDEXPANSION:
>
>    /tmp/foo.o:
>
>    %.o: $$(addsuffix /%.c,foo bar) foo.h ; @echo $^
>
>    $ make
>    /tmp/foo/foo.c /tmp/bar/foo.c foo.h

I was looking in the wrong place.  The pattern_search routine in 
implicit.c appears to be the culprit.

It needs to be able to handle paths in both VMS syntax or in Unix syntax 
under VMS.

It also does not appear to be properly handling VMS syntax either.

Fixing it to handle the Unix syntax should be easy.

Fixing it to handle the VMS syntax correctly is far more complex, as VMS 
syntax has changed since the original port of Make was done, and it does 
not look like the original code was handling the older syntax correctly.

Regards,
-John